OCCASIONAL PAPERS OF THE MUSEUM OF ZOOLOGY UNIVERSITY OF MICHIGAN

Transcription

1 OCCASIONAL PAPERS OF THE MUSEUM OF ZOOLOGY UNIVERSITY OF MICHIGAN TI-II<EI3 NEW SPECIES OF TIIE GENTJS EHAI)Ii\'AEA b1120m (f IJArl'EMATiA BY L. C. STUART AND JOSEITI It. ~SAILEP REPENT c~)lle~ti~~g ill the cel~tral il~o~uitain ranges oi' Gn;~tcinala by h'li.. l<;~yinoncl Stadelnian ol' tl~e TJliited States Dep;n-liiicilt ol' Agricnlture a~ld by the senior a~~lhor has I-esnltctl ill the capt~~re ol' 6 spec.ii~~eils of the gcin~s lihnrll~lczca. Though it 111;ly seein o\rercritical to assign these G sl)cc~iinenr to 3 spec.ies, the jniiiol- anthorl has already noted that tlrc tciitlcilcy of Illis gcllus to itlvi~de high altitudes has n~lt1ouk)tctlly c*o~llribr~lcltl to the inultiplic~ation of spccits 01 Iiinited distrll)~~lioil. Tlre Ire\\ I'oi.ms origliintetl, moreol c.r, Croiil tlie Ilcl-petolog rc;~ll y poorly ]<iron 11 irio~lilt~xir~s ol thc Alta Verapal: aiicl f roll1 the \\~l~ol lg 11 lllaro~~ll Sierra clc 10s C~rclluinat ai~cs. All albl.w~lgeincnt of the Mesicail specics ol' IZIicrdr ncrca has recc~il11jr been preselltrd by tire jnllior. b11t tliere i\ still 111uc211 to he clesiretl iir our lrrlo\\ lctlge of the Central Aine~-~e;iii Sorms. The first 2 species tlescribecl below clarify, lo seine cxtcilt, lhr lacir~~~j~rrcrns-go(inrn~~~~ relatiollsl~ips s~~ggestcd ill 1 Jo\cl)l~ R. Bmlcy, "Tllc h!ie\~c:ni Sllnlres of tlie Gcilus BharT~naca," Occ. I'ng,c~\ 1Wt1\. Zool. U~[iu. Jfztl~., 412 (1940) Ibrtt. 3 Tllor~gll Cruiitlici uscd :I doul)lc "II" m tllc o11g1n:rl clcscliption of tlns Eoiir~, wc bc~l~cvc Il1:it it \\:as :r lap~\ rtala?nz, x.: lic sl~cllcd Codln:111's n:r~nc COll cc tly scvcl'll llllcs 1'~tcl.

2 2 Xtz~art and Bailey OCC. Papers the junior author's paper.4 Until Nicaragua and Honduras are better kno~vn, few suggestions can be made as to relationships betnreen northern and southern Middle American forms. The first species to be named is dedicated to Mrs. R. W. Hempstead of Cob&n, Alta Verapaz, Guatemala, to whom the senior a~lthor is greatly indebted for the many courtesies she extended to him during his investigations in the Alta Verapaz. It mag' be known as Rhadinaea hempsteadae, new species Ho~o~~~~.-University of Michigan Museum of Zoology (U.M.N.Z.) No , an adult male collected by L. C. Stuart in the cloud-forest zone (en feet) above Finca Chichdn, Alta Verapaz, Guatemala,"fay 26, PARATYPE.-U.1l.M.Z. NO , from the same locality as the holotype, though from a higher altitude. D~~a~osrs.-A Rhadinaea of the laclzrynzans-godnzani group, with 19 rows of dorsal scales, abdominals, and subcaudals. A dorsal pattern of a narrow, dark, middorsal stripe, and a light lateral stripe on either side oil the fifth dorsal scale row. DESCRIPTION OF THE TYPE.--H~~~ scales normal; internasal suture 4 as long as the prefrontal snture; frontal very slightly longer than its distance from the end of the snout and shorter than the parietals. Supralabials, 8, fourth and fifth entering the orbit; infralabials, 8 on one side, 9 on the other, 5 i11 contact with the anterior chin shields, whicll are shorter than the posterior ones; oculars 1-2; temporals Dorsal scales in 19 rows on the body ; abdominals, 163 ; subcaudals, ; anal divided. Maxillary teeth, 21, increasing in size posteriorly, the last 2 strongly enlarged but not separated from the others by a diastema. Body length, 266 mm. ; tail length, 128 mm. Dorsally the ground color is brown. There is a darker middorsal stripe, 1 scale row in width. Laterally, the center of 4 Op. cit., pp Fillca CIlichBn is in the pine zone of the Alta Verapaz about 10 km. (straight line) south and slightly east of Cobkn, Alta Verapaz.

3 Three New Species of Rhadinaea each scale in the fifth row is lighter, to produce a light lateral stripe, bordering a dark stripe which is situated on the lower 4 of scale row 5, on all of scale row 4, and on the upper 4 of scale row 3. The color of the lower 4 of scale row 3 and the upper half of scale row 2 is somewhat lighter than the ground color. The outer edge of all the ventral scutes is somewhat darkened, and gives the appearance of a ventrolateral stripe. The head is dark brown dorsally and laterally. The supralabials each possess a distinct, large, white spot, and there is a white bar which extends from the lower, posterior edge of the eye to the angle of the mouth. As in stadel??zani (p. 4) there is an indication of a light rostral border. The entire uiidersurfaces are white except for brown fleckings on the anterior infralabials aiid for the brown outer edges of the ventral scutes. The senior author's field notes indicate that in life the dorsal ground color was reddish brown. Lighter flecks on scale rows 2 and 3 and oil scale row 5 suggested 2 light lateral stripes. The labial spots aiid the eye-mouth bar were bright yellow. The undersurfaces were lemon-yellow. VARIATION.-T~~ paratype, a male, is similar to the holotype in all respects. It has 166 abdominals, 93 subcaudals, and only 18 maxillary teeth. The dorsal ground color is grayish brown, or grayish blue where the stratunz corneum is lacking. The middorsal dark stripe is very faint anteriorly and absent posteriorly. The fifth row of dorsal scales is white-spotted throughout its length, and scale row 2 is somewhat lighter anteriorly. Posteriorly the specimen is unicolor except for the light stripe on scale row 5. The eye-mouth bar is absent on one side and somewhat obscure on the other. A headless specimen from the cloud forest above Finca Volcin, some 45 km, due east of CobBn (in a straight line), is reddish brown above, with the dark middorsal stripe and light lateral stripe anteriorly. When found, freshly killed, it mas coffee-brown above and lemon-yellow beneath. RAN~E.--Probably restricted to the cloud forest between the Cahabdn and Polochic rivers in the main mountain mass of the Alta Verapaz, Guatemala.

4 4 Stuart and Bailey Oce. Papers I~m~~s.--This species is appareiitly a secretive inhabitant of the cloud-forest zone, the lower limit of which in the Alta Verapaz is at about 5000 feet. The holotype was found beneath a log in a cleared milpa, and the paratype was talcen from beneath a pile of wood chips in the cloud forest. The headless specimen was fouiid dead on a trail in the cloud-forest zone. Fronz the stomach of the holotype was taken a specimen of Oedzpus helmrichi Schmidt. Inasmuch as the senior author has never found this salamander on the ground, and since E. hempsteadae is a gracefully attenuate fonn, it is suggested that it may, possibly at night, climb to some extent. GELATIONSHIP.-This species is believed to represent a peripheral offshoot of stadelmani. Its eye-mouth bar, light-bordered rostral, and tendelicy towards 3 dark anterior stripes certainly place it in the godnzani-lachrymans series. It has apparently invaded the Alta Verapaz from the west, a history similar to that of several other Verapaz species. For Mr. ltaymond Stadelman, xvhose collecting in Iloiiduras aiid Guatemala has made him well known to herpetologists, we name Rhadinaea sfadelmani, new species HOLOTYPE.-U.M.M.Z. No , a juvenile fernale collected by Mr. Eaymoild Stadelnlaii at Todos Santos, IIuelinetenango, Guatemala, at 8000 feet altitude, November, PARATYPE.-U.M.M.Z. NO from the type locality. ~r~a~os~s.--a Iilzadznaea of the lac~zrymans-godmani series, with 19 rows of dorsal scales, abdominals, and subcaudals. A dorsal pattern of a narrow, dark, middorsal stripe, and a lateral stripe on the common borders of the third aiid fourth scale rows; posteriorly another stripe on the common borders of the sixth and seventh scale rows. DESCRIPTION OF LIOLOTYPE.--II~~~ scales normal ; internasal suture less than ', the length of the prefrontal suture ; frontal considerably longer than its distance from the end of the snout an$ shorter than the parietals. Supralabials, 8, fourth and fifth entering the orbit; infralabials, 8, 4 in contact with the

5 No. 442 Three New Species of Rhadinaea 5 anterior chin shields, which are shorter than the posterior ones ; oculars, 1-2; temporals, Dorsal scales in 19 rows over the length of the body ; abdominals, 179 ; subcaudals, ; anal divided. Maxillary teeth, 17, increasing in size posteriorly and lacliing a diastema. Body length, 130 min. ; tail length, 43 inm. Dorsally the ground color is brown. The central of the middorsal row of scales is darkened to produce a narrow, dark, middorsal stripe. Laterally the upper of scale row 3 and the lower 4 of scale row 4 are also darkened to form a somewhat broader, clarlr, lateral stripe. Another faint stripe appears on the posterior third of the body as a result of the darlreniiig of the upper $ axid lower : of scale rows 6 axid 7 respectively. In this same region the lower lateral stripe is somewhat obscured because scale rows 1 and 2 are darker. The brown gro~lnd color of the dorsuin infringes onto the outer edges of all the ventral scutes, which are otherwise white. llhe ground color of the head is dark brown, both above aiid laterally. There is an irregular spot of white along the posterior edges of the frontal, and the rostra1 and adjacent scales, anterior-ly, are bordered with white. Tlic snpralabials each coillain a sharply denlarliecl white spot, and a light stripe extends f~*om the infcroposterior border of the orbit to the aiigle ol the th. There is a large white spot laterally on either side of the nape. The inferior surfaces are immaculate white, with the exceptioil of the brown spotting 011 the infralabials ant1 the brown niarkiligs on the outer ends of the ve~itral scldcs. VARIATION.-T~~ paratype, also a juvenile, with a crushed liecli aiid partially crashed head, is esseiitially lilie the holotype. It has 184 abdoininals, 77 subcaudals, and 19 dorsal scale rows anteriorly, which are rednccd to 17 posteriorly through the fusion of the paravertebral (ninth) rox7s with the eiglith row on either side. The sole difference iii pattern is to be foulid in thc greater clarity of the superior lateral stripe on the posterior part of the body. Rh~a~.-Tho~gh B~iown oiily from the type locality, it is not

6 6 fltuart and Bailey Oce. Papers improbable that this species is endenlie to the Sierra de 10s Cuchunzatanes of IIuehuetenaago and El Quichk. HABITS.--TJ~~ know nothing of the habits of this species other than that it was collected in the pine zone on the slopes of the Cuehumatfin niountains. RELATIONSHIP.-In his recent paper, the junior autl~or,~ although uncertain as to the position of Blzadi~zaea lacizry?~zans (Cope), suggests that it is a relative of R. godnza~zi (Gunther). That such is the case is now more evident on the basis of this new form. In pattern it is almost identical ~i7ith godnzani, with the exception that there is an additional pair of lateral stripes posteriorly. Its posterior pattern is also close to that of lackrymans. With its 19 dorsal scale rows, it is internlediate between godnzani and lachry+~za?zs, which possess 21 and 17 dorsal scale rows respectively. Though stadelrtzani is geographically not exactly annectent between godnzani and lacl~rynzans, it is sufficiently close to offer no difficulties to being regarded as ail intermediate form. An account of the group relationships will be presented below. RELATIONSHIPS OF TI-IE Lacanr'nr~ixs ~~ou~.-the laclzrynzafisgodnza~zi relationship previously suggested by the junior author7 is, through the discosrery of the above-described species, greatly clarified and now perniits the outline of another natural group of the coniples genus Rhadzkaea. Two species of Rhadinaea, godnzani and serperaster, have stood out from the great mass of Middle American forms by virtue of their distinctive bead pattern (described below), their possession of more than 17 rows of dorsal scales, and by an absence of a diastema in the maxillary dentition and of anal ridges in niature males. To these 2 species ve nlay now add stadel??zanis and Izempsteadae, which are similarly characterized. These 4 species are linked to the more usual Rhadinaea types G LOG. tit. 7 01). cit., pp Since we have not had access to adult specimens of stadelmani me are unable to state definitely that anal ridges are a diagnostic group character.

7 NO. 442 Three Ncw Species of Rhadinaea 7 by R. Zacl~rynzans, which resembles the former in its lack of anal ridges in aclult males and in its head pattern, and approaches tlie latter i11 possessing but 17 rows of dorsal scales ancl a cliastema. Thougll lnchrywza7zs is a typical ailneetent form, we prefer to place it in a gronp with the other 4 species because of its head pattern and lack of anal riclges and because, as will be sl~o~vn, it represents an end form in number of dorsal scales. These 5 species, ~ilhich we may call the lachry~.lza~zs group, inhabit the high rnountaiiis of Chiapas and Gnatelllala, except sel-peraster, which is Bno~vn froin Costa Rica. ddditioiial nleiiibers of the group undoubtedly amit discovery i11 the higher altitudes of the intervening mountains of Honduras and Nicar a g ua. In the number of dorsal scale rows R. godnzani stands at one end of the series with 21 aiid R. lachrynzalzs at the other end, with but 17. Between these extremes n7e find R. serperaster, with 21 or 19, aiid R. l~enzpsteadae and R. stadelnza?zi, with 19. The paratype of stadelnzani sho~s a dorsal reduction of from 19 to 17 rows posteriorly, o~i-ing to a fusion of the ninth (paravertebral) and eighth scale rows on either side. Reduction in number of dorsal scale rows is, aside from the abo~e instance, unknown in Middle American Rhadinaea, but this same reduction is found in tlie related genera Dindophis aiid Coniopkn~zes, so that its occurreilce here is not surprising. Our interpretation of dorsal pattern is based on the assumption that increase or reduction in dorsal scutellation is centered around the paravertebrals. In the dorsal pattern of this group, as in tlie hiesican species, there is a tremendous amonnt of interspecific variation. Figure 1 is a diagrammatic representation of the pattern in the several species. Two sets of stripes occur in the group, the one which we refer to as the major or primary stripes (solid in the diagram) and the niinor or secondary stripes (stippled in the diagram). A11 of the species of the lacl~ry~~zans group have a major stripe, which varies in width, on the middorsal row. The common borders of the third and fourth ro~vs like- 1i7ise bear a major stripe, which in henzpsteadae is widened

8 8 Stuart a?zd Bailey ~cc. Papers above to include part of the fifth rov and ia lacl~ryma?zs is widened below to include a part of the second row. Secondary stripes occur 011 the outer edges of the ventral scutes (occasionally overlapping slightly on the lower half of the first scale row) and 011 the collilnoli borders of the sixth and seventh scale rows. Ill R, serperaster this latter is midened above to include part of the eighth scale row. As in scutellatioa, R. stadel?~ze.izi stands half way between R. godnta~zi SERPERASTER STADELMANI HEMPSTEADAE LACHRYMANS FIG. of the Diagrammatic represell 'ry?nans group of Rhadj f species and R. lacl~ry?71ans. The pattern is most colnples in serperaster and least so in Izenzpsteadae. In lacl~rynzans there is a. distinct telidency for the stripes to clarkeii with age, eve11 to the extent of the 2 secondary stripes becoming as intense as the primary ones. To what extent ontogenetic changes in pattern occur ia the other species we are not preparecl to say. The distinctive head pattern ~vhich serves to tie this group together is characterized by a light border around the posterior

9 NO. 442 il'lzree New Species of Rhadinaea 9 edge of the rostral, promineilt light spots on the supralabials, a light bar from the eye to the corner of the mouth, and the abse~ice of a light temporal stripe behind the eye. This pattern is a diagnostic character of the group. Thc abseilce of a diastema in the niaxillary dentition i11 species of Rlzaclinnea with more than 17 rows of dorsal scales is possibly a "primitive" character and way, in coi~ibinatioii with the higher scale formula, indicate an ancestral position i11 the genus. The diacranteriaii clentition, on the other haiicl, is much conlriioiier than tlie syiicra~lteria~l type iv the Xenodontiilae and, with the increased number of dorsal scales, may well irldicate specializatioli on the part of the lachrynzans group. In whiclievcr direction the series inay progress, it is iiidicated that the relationships of the group are as follows: unkiscovcred species 9 I It is vlith sorne hesitation that we describe a second species lrom the Alta Verapaz. Onr type proves to be a juvenile male, but since, even allowi~ig for telc juvenile pattern, we are unable lo allocate it to any lriiown species group, we feel that attention should be called to it through a description. We name it Rhadinaea veraepacis, new species IIOLOTYPE.-U.M.M.Z. No , a juvenile male collected i11 the pine zone (5100 Et.) at Filica Chichdn, Alta Verapaz, Guatemala, by L. C. Stnart, May 17, Draa~osrs.-A ZZlzadinaca of obscure relationships, possessing 17 dorsal scale rows, 145 abdominals, and 80 snbcaudals. A dorsal pattern of a middorsal dark stripe and aiiotller distinct, dark, light-centered lateral stripe on either side. DESCRIPTION OF IIOLOTYPE.-Head scales ~iormal ; internasal suture about as long as prefro~ital suture; frontal about 4 as

10 Stz~art and Bailey OK. Papers long as tlie parietals and much longer than its distance from the end of tlre snout. Slipralabials 8, fourth and fifth entering the orbit; infralabials, 8, 4 in contact with the anterior chin shields, ~vliich are longer than the posterior ones ; oculars 1-2 ; temporals 1 I 2. Dorsal scales in 17 rows over the length of the body ; abdominals, 145 ; subcaadals, 80 i- 1 ; anal divided. Maxillary teeth, 13, increasing in size posteriorly and wit11 a diastema anterior to the last 2. Body length, 111 min. ; tail length, 47 mm. The dorsal grollnd color is light brown. The iniddorsal scale row is much darker brown, with ininute white flecks, and it prodnces a clarlr middorsal stripe. Laterally, scale row 4 and the upper ancl the lower : of scale rows 3 and 5 are similarly darliened, and a broad lateral stripe is thus produced on either side. The center of each scale in scale row 4 coiitains a light fleck, so that actually the broad lateral stripe appears as 2 narrower stripes in close proximity. A solnewhat lighter lateral stripe with a similar light center appears on scale row 6 and tlie edges 01 the acljacent scale rows. Other bands, darker than the ground color but lighter than the middorsal or niain lateral stripes, occur on the lower : of scale row 1 and the outer edges of the ventral scntes, anothcr on the upper and the lower!% of scale rows 1 and 2, and yet another on the upper :3 and the lower of scale rows 2 and 3. The dorsal surpace and sides of the heacl are dark broxvii. The posterior portion 01 the frontal is strongly punctated with white, and the rostra1 is ~ilhite-bordered above and laterally. The sllpralabials are brown, each of the first 7 with a large white spot. The infralabials, chin, and throat are white, slightly flecked with brown anteriorly. The rclnaining undersurfaces are ~vhite except for the brown lateral edges of the ventral scutes. There is a white vertical band laterally on either side of the nape. Rh~~~.-Though known only froin the type locality, this species may prove to inhabit the mountains of the Alta Verapaz below R. hempsteadae.

11 NO. 442 Threc New Species of Rltadinaea 11 11~31~s.-Nothing is lrno~vii of the habits of this species. The holotype was foulid ensnared in a spider-web in the hacienda at Finca ChichPn, which lies in the heart of the pine zone of the Alta Verapaz. RELATIONSEIIPS.-The relatioiiships of Rhadinaea veraepacis are, at the present, obscure. Structurally, in having 17 dorsal scale rows and a distinct diastema iii the maxillary dentition, it is closer to the other groups of Rl~adinaea than it is to the lachrl~mans complex, which it resembles some~vhat in head pattern. Though it laclcs the light eye-mo~~th bar, it does possess a light rostra1 border aid prominelit light spots on the supralabials. Its tciidency toward possession of light streaks aloiig the ceiiters of the dorsal scale ro~17s may iiidicate a paclzyurafulviceps-lateristriga relatio~iship to the south. The latter species present a body pattern which is t~nifornily dark, brolien by 5 light strealis on certain of the scale rows. It is possible that this type of pattern forills one evolntioiiary extreme while the predominately light dorstllii with accpssory dark stripes, as in the Zachrymans group, presents the opposite encl. ACKNOWLEDGMENT The senior author wishes to ackiiowledge fiiiaiicial aid received froin the Horace H. Raclcham School of Graduate Studies, aiicl thc Baird Eiiclowliieiit Fund, University of Michigan, which made possible his investigations in the Alta Verapaz ancl in the Sierra cle 10s Cuchumatasies.

12